Search This Blog

Monday, November 14, 2011

Here we go!

I’ve been busy, but I’ve somehow found time to work on LiteBoard. Oh right…what is LiteBoard?

Now Presenting LiteBoard:
But first, some history: I want a light*, fast way to get around campus, which is the same premise and idea that evolved into ELB. Unfortunately, ELB turned out to be more of a prototype and a learning experience. Now I want a light*, fast, and polished product to get around campus. Sure, there are commercially (and cheaper) available options, such as a bicycle (can’t pick it up and take it to class…also not original and would get stolen), roller blades (no way…I’d kill myself on those), kick scooters (lammeee), electric scooters (lammeerrr), hub motor scooters (so not original)…but I wanted a longboard. So I came up with the EHB concept, and started acquiring parts for it. But once I did a weight build-up, I realized that EHB wouldn’t actually be much lighter than ELB, so I suspended work on it. I’ve been riding around on the mountainboard I bought for EHB for awhile now…I have to say, it is a great way to get around. It’s just slow: I can only maintain about 10mph.

LiteBoard is the V3 concept that I came up with after completing ELB. V2 is/was EHB, but even though it would address most of the problems ELB has, it didn’t address one very important one: weight. ELB weighs 40lbs. EHB is predicted to weigh 35lbs. Liteboard: 25 lbs.

Not enough time to do a rendering=screen shot.
Lite will be a 2WD hub motor powered electric longboard (actually a longboard this time, and not a mountainboard). Each motor will be ~1100W, producing ~3.9Nm of torque each. Lite’s estimated top speed will be ~25mph. The basis of Lite is this longboard:


Pending cool grip-tape design

Randal II 50deg 180mm trucks, ABEC11 97mm Flywheels, and a bamboo/maple deck I got off of ebay (it was incredibly cheap, yet very high quality. ebay=good source for longboard parts). The deck needed some modifications to use the massive 97mm Flywheels:

Initial cutout lines. 
1st try: didn't remove enough.
Second try: Now the wheels don't bite the deck when I turn.
I think it looks awesome...
Why did I need 97mm Flywheels? …

Detailed Motor design:


I couldn't get a good angle of the internals. The axle extension is not round, but keyed to fit with the hub that is pressed into the stator.
Since ELB ended up with about twice as much torque as I’d ever need, I decided to shoot for the same amount of torque/motor for LiteBoard, but cut the number of motors in half. I followed the same design process I used to design ELB's and EHB's motors: Start with simple calculations, move on to SolidWorks CAD drawings, import DXF's into FEMM, modify CAD, repeat...

FEMM pre-diameter shrinking.
The motor underwent one large design change as a result of FEMM. Turns out that pretty much the whole stator was saturated, so I wasting a ton of energy. I shrunk the magnets from 3mm to 2mm thick, and shrank the whole OD of the motor by 4 mm, and only took a .1Nm hit in the process.

Estimated Motor Specs:
Specification
Value
# Phases
3
Power
1100W
Voltage
6S Lipo
Max current
50A+
Torque (FEMM)
3.9Nm
No-load speed
30mph
Stator dia
55mm
Stator length
35mm
Teeth
18
Turns/tooth
4
Rotor OD
70mm
Magnet thickness
2mm
Magnet grade
N40
# Poles
20
Motor width
60mm
Motor OD
100mm

The high current, low voltage method was picked in order to use HobbyKing’s 80A car ESC, which is about 1/3 of the cost of an equivalent Kelly Controller, and about ¼ the size. It also has a sensored option that I plan on using. I will likely get some pretty serious I^2R losses, but I’ll try to keep wires as short as possible and pack the stator with multiple parallel strands of magnet wire (thinking 22Ga). 8 parallel 22ga stands has same resistance/m as one 13ga strand…idk, I’ll see how many strands I can fit.
The stators are Scorpion Power Systems 5535 18T stators. They come in their S5535 motor series, which are about $400 a piece…which is why I bought the stators directly from them again ($140 for both, free DHL express shipping). Yet again, they are very high quality:



I officially love these guys' work.
I decided to bite the bullet and go with BigBlueSaw for waterjetting my rotor plates. Turns out that 11*.125” =35mm to within a fraction of a mm…I got lucky there. 22 plates will run my $250…ouch! But I did the math, and if I spent the amount of time working that it would take me to machine the rotors, I would come out ahead if I just have them cut for me (and waterjetting them here is a no-go…it’s freakin’ expensive).

Batteries: While I thought about trying A123 cylindrical cells for this one (specifically a 7S7P pack made of their 26650 cells), it turns out that it won’t fit under my deck. So I went with lipo again. Specifically, I’m going to buy 2 more 6S 5000mAh 25C packs from HobbyKing, and using the leftover packs from ELB, I’m going to make my own, slim 6S4P pack (22.2V, 20Ah).

Control: At first I’ll use a $15 hobbyking 2Ch. 2.4GHz system. But I’m planning on doing something like this eventually.

Tires: So my last post suggested a Colson 4”x7/8” rubber wheel was my best bet. But that was before I shrunk the OD of the motor from 74mm to 70mm, which puts it out of the range of the 7/8” one. So I’m going to use two 4”x1.25” ones.

From left to right: 4x7/8 turned out, 4x7/8, 4x1.25, 4x1.5,4x2

Note the huge void.

After digging a bit deeper: MORE huge voids.
 Next I tried the 1.5" wide one:

Only 1" wide of plastic inside (after turning out visible part of hub).  Guess I'll be using the 1.25" wheels (more plastic for press fit).
I was able to cut the rubber on these.
Again: another void...

Here's a table of 4" Colson rubber wheel data for your reference.

Safety: I’ll have the same setup as I did on ELB. A contactor sitting in the main battery line that is actuated by a magnetic reed-switch from a magnet strapped to my shoe. This will act as a fail-proof way to make sure that if I’m not on the board (anymore…), it doesn’t keep going. I’m also planning on wearing a helmet with it.
The ultimate goal is have a very sleek, finished product that achieves all of my criteria (safe, light, reliable, fast, etc…)
Anyways, that’s the plan so far. I should be able to get a lot of it done over IAP…we’ll see.

Oh, and to leave you with a cliffhanger: it’s called “Lite” for more than one reason ;) .